Random forest algorithm for algorithm for prediction of high school science students acceptance snmptn based on students assesment report

Abstract

National Selection for State University (SNMPTN) is one of the selection
lines for admission of new students in Indonesia to enter State Universities by
invitation. Report card grades are one component of the assessment of
admission of new students to enter state universities on this pathway. The
difference in standards between universities in determining the admission of
SNMPTN applicants, causing the need to predict based on several related
factors. This research uses data mining techniques with Random forest
algorithm. From the results of research that has been done, it was found that
the Random Forest algorithm can be used to predict students who are accepted
at SNMPTN based on report card grades, obtained from the results of the
classification process with the student report card report survey dataset
received by SNMPTN, This is indicated by the accuracy, precision, and recall
values of 93%. Optimization of the random forest algorithm using the
oversampling technique with the SMOTE method can improve the classifier's
performance due to the imbalanced class problem.